Crux

Mathematicorum
Published by the Canadian Mathematical Society.

http://crux.math.ca/

The Back Files

The CMS is pleased to offer free access to its back file of all
issues of Crux as a service for the greater mathematical
community in Canada and beyond.

Journal title history:

➢ The first 32 issues, from Vol. 1, No. 1 (March 1975) to
Vol. 4, No.2 (February 1978) were published under the
name EUREKA.
➢ Issues from Vol. 4, No. 3 (March 1978) to Vol. 22, No.
8 (December 1996) were published under the name
Crux Mathematicorum.
➢ Issues from Vol 23., No. 1 (February 1997) to Vol. 37,
No. 8 (December 2011) were published under the
name Crux Mathematicorum with Mathematical
Mayhem.
➢ Issues since Vol. 38, No. 1 (January 2012) are
published under the name Crux Mathematicorum.
 E U R E K A

Vol. 1, No. 5

July 1975

Sponsored By
Carleton-Ottawa Mathematics Association Mathematique d'Ottawa-Carleton
Published by Algonquin College
Send all communications to the editor: Leo Sauve, Math-Architecture,
Algonquin College, Col. By Campus, 281 Echo Drive, Ottawa, Ont., K1S 5G2.

NOTES FROM THE EDITOR

1, One hundred seventy-one copies of EUREKA No. 4 were mailed a few days ago,
most of them going to the Ottawa region, a few dozen elsewhere in Ontario,
a few in Quebec, and one to the United States. New subscribers are welcome. Get
your mathematically-minded friends to send the editor their name, address, and place
of employment. He will add them to the mailing list and send them back issues. .
2, Even more than additional names on the mailing list, the editor would like
to see greater participation by existing readers, in the form of solutions
to problem proposals. To make his job easier, he would like to remind you that solu-
tions, if not typewritten, should be neatly handwritten on signed, separate sheets,
in a form suitable for publication.
The expression "in a form suitable for publication" means that a handful of
mathematical symbols sprinkled on a page does not constitute a proof. The words and
mathematical symbols appearing in your solutions should collectively form recogniza-
ble, grammatical, properly punctuated English (or French) sentences, accompanied by
neatly drawn diagrams where appropriate. Careful phrasing is conducive to clear
thinking. It is not easy to write a proof in which you are convinced that every last
comma is exactly where it should be; but what is worth doing is worth doing well.
Regardless of anything said in the last paragraph, the editor would like
to encourage readers who are young or inexperienced in mathematical writing to send
in their solutions in spite of a fancied lack of elegance in their presentation. After
all,in launching EUREKA, one of the principal aims of the founding members was preci-
sely to help these people in their struggle to achieve mathematical maturity. So do
the best you can and send your solutions in; what is worth doing is worth doing badly.
3, The mathematical elite in our community, the university professors and gra-
duate students, can help a great deal by submitting interesting articles and
model solutions to the more difficult problems. They will themselves benefit in the end,
for increasingly competent high school teachers will eventually result in increasingly
competent high school graduates and university students.
if, EUREKA will normally be published ten times a year, every month except July
and August, when most of its readers (and the editor) are on holidays. Since
the first issue of EUREKA was in March 1975, it will this year, exceptionally,be pu-
blished in July and August, so that Volume 1 (1975) will have a full complement of ten
issues.
 - 38 -
In the future, at the request of several readers, solutions will be pub-
lished three months, instead of two, after the problems are proposed. In order to
make this possible, the August issue will contain new problem proposals but no solu-
tions.
5, The editor hopes that readers enjoyed his report on Martin Gardner's hoaxes
in the last issue of EUREKA. Here, as promised, is further information on
the matter:
i) Gardner has promised to give, in the July issue of Scientific American,
a report on the public reaction to his hoaxes.
ii)Dr. John D. Brillhart, the presumed alter ego of John Brillo, is a widely
known number theorist from the University of Arizona. Among his discoveries is the
factorization of the Mersenne number 2 1 0 3 — 1. It is

2 1 0 3 - 1 = 2550183799 x 3976656429941438590393.

He has also proved that 2 1 3 1 - 1 is a product of two primes, one of which is 263.
And that is no April Fool's joke.

P R O B L E M S - - P R O B L E M E S
Problem proposals, preferably accompagnied by a solution, should he sent to
the editor, whose address appears on page 37.
For the problems given below, solutions, if available, will appear in EUREKA
No. 8, to be published around October IS, 1975. To facilitate their consideration,
your solutions, typewritten or neatly handwritten on signed, separate sheets, should
be mailed to the editor no later than October 1, 197S.

41, Propose par Leo Sauve, College Algonquin.

Ayant donne log B 3 = p e t log 3 5 = q, exprimer logioS e t l o g 1 0 6 en f o n c t i o n
de p e t q.

42. Proposed by Viktors Linis, University of Ottawa.

Find the area of a quadrilateral as a function of its four sides, given that
the sums of opposite angles are equal.
45, Proposed by Andre~ Bourbeau, Ecole Secondaire Garneau.
In a 3 x 3 matrix, the entries a., are randomly selected integers such that
o < a., i 9. Find the probability that
I'd
(a) the three-digit numbers formed by each row will be divisible by 11;
(b) the three-digit numbers formed by each row and each column will be divi-
sible by 11.

44. Proposed by Viktors Linis, University of Ottawa.

Construct a square ABCD given its centre and any two points M and N on its
two sides BC and CD respectively.
 - 39 -

/ | 5 , Proposed by H.G. Vworschak, Algonquin College.

Find c o n s t a n t s A, B, C, D, p, q such t h a t
A[x - p)2 + B(x - q)2 = Sx2 + 8x + 14,
2 2
C{x - p) + D(x - q) = x2+ 10a: + 17.

i)6, Proposed by F.G.B. Maskell, Algonquin College.

I f p , p , p are the altitudes of a triangle and r is the radius of i t s
inscribed c i r c l e , show that
1 + 1 + 1 - 1
Pi Pi

i\J, Propose" par Jacques SauvS, etudiant, Universite d'Ottawa.

°° k2
Si a > 1, calculer la somme de la serie £ —r.
fc=i a
48 • Propose par L&o SauvS, College Algonquin.
La fonction f:R*R est definie par les relations
i
f{x) = 2 + Sl'na;COS — , Si x ^ 0,

f(0) = 2.

Pour tout entier n > 1, on considere l'integrale

T =
n \ \ { n + {
n ~ n
K { x ) }
^x)dx>
~n

ou X desiqne la fonction caracteris.tique de l ' i n t e r v a l l e , — . Calculer I en

n [_ n nj n
fonction de n et en deduire la valeur de lim I .
n
n->- ^
i|9 i Proposed by E.G. Vworschak, Algonquin College.
The s e r i e s
1 _ 1 + 1 _ JL + + 1 1 i
5 7 11 6n-5 6n-1 "" '
clearly converges. Find i t s sum.

5 0 i Proposed by John Thomas, Digital Methods Ltd.

I found the following fascinating two-part problem in Martin Eisen's Introduc-
tion to Mathematical Probability Theory (Prentice-Hall 1969). Information as to i t s
origin and history would be appreciated.
{&)Show that 2 can begin with any sequence of digits.
(b)Let N be an r-digit number. What is the probability that the first r digits
n
of 2 represent N?
 -40 -

S O L U T I O N S
21, Proposed by H.G. Dwovschak, Algonquin College.
What single standard mathematical symbol can be used with the digits 2
and 3 to make a number greater than 2 but less than 3?
Solutions^ by Keith Bateman, Algonquin College; Don Hull, Hillcrest High School;
Andre Ladouceur, Ecole Secondaire De La Salle; Viktors Linis, University of Ottawa;
F.G.B. Maskell, Algonquin College; Leo SauvS, College Algonquin; John Thomas, Digital
Methods Ltd; and the proposer.
Each of the above solvers gave one or more of the following solutions:
2.3, ^ = 2.83, 3sin2 = 2.73, 2arctan 3 = 2.50, 3.612 = 2.08,
2 In 3 = 2.20, CSC 3 2 = 2.43.
Comment by G.D. Kaye, Department of National Defence.
Readers of Kurt Vonnegut's science fiction will recall a similar message con-
veyed across the universe by a robot. In robot language, it signified "Greetings".
Thanks, Mr. Kaye, and 2.3 to you too. (Ed.)

22 1 Proposed by H.G. Dworschak, Algonquin College.

Numbers are written on little paper squares as shown in the figure.

s LH \n
s Q \n 8
Show how to make the sums of the two rows equal by moving just two of the pieces.
I .Solution de Nicole Trudel-Marion.
On renverse le 9 pour en faire un 6 et on l'echange ensuite avec le 8.
La somme de chaque rangee devient alors 18.
C'est un probleme parfait pour les femmes de mathematiciens!
Editor's comment. Nicole Trudel-Marion is the recent bride of Jacques Marion,
a graduate student in mathematics at the University of Ottawa.
II. Solution by F.G.B. Maskell, Algonquin College.
Move the 1 to the bottom row and raise the 3 slightly. The result is

2 + 7 + 9 = 18,
3
1 .+ 4 + 5 + 8 = 18.
Also solved by John Hayes; G.D. Kaye, Department of National Defence; Andre"
Ladouceur, Ecole Secondaire De La Salle; and the proposer.
 - ill -

2 3 • Propose par L&o Sauve, College Algonquin. b

Determiner s ' i l existe une suite {u n } d'entiers nature!s t e l l e que, pour

n = 1, 2, 3,..., on a i t

2*" < 2n + 1 < 2 1 + " «

I . Solution de John Thomas, Digital Methods Ltd.

La fonction logarithme de base 2 etant strictement croissante, la relation
proposee equivaut a la suivante:

u < log 2 (2rc + 1) < 1 + u .

n ^"- ' n
S'il existe une telle suite, 1'entier u est necessairement le plus grand entier in-
ferieur a log2(2n + 1), que 1'on note, comme d'habitude,

un =[jog2(2n + 1)].

L'inegalite est stricte, car 2 M " est un nombre pair.

On voit facilement que cette suite convient; elle est par consequent la seule.
II. Solution by G.V. Kaye, Department of National Defence.
For n = 1, 2, 3,..., let u be the number of digits in the binary represent-
ation of n. If

n = ixx x ... x (u digits),

n
where x is o or 1, then
2n + \ = ^xx x ... x\ (it + 1 digits).

This is obviously greater than 1 ooo ... o [un + 1 digits), which is 2"", and less than
1000... 00 [u + 2 digits), which is 2 " n + 1 , so that
2Un < 2n + 1 < 2 1 + V

Editor's comment.
The uniqueness of the result, together with the different representations of
i t given in solutions I and I I , gives
\}og2{2n + 1)J = number of digits in the binary representation of n,
which is otherwise not immediately obvious.
Also solved by Andre Ladouoeur, Ecole Secondaive De ha Salle; Viktors Linis,
University of Ottawa; F.G.B. Haskell, Algonquin College; and the proposer.
 - 42 -

24, Proposed by Viktors hints, University of Ottawa.

A paper triangle has base 6 cm and height 2 cm. Show that by three or fewer
cuts the pieces can cover a cube of edge 1 cm.
Solution by G.D. Kaye, Department of National Defence.
The triangle can be dissected, by two or three cuts, and the pieces rear-
ranged into two 3 cm x 1 cm strips which will cover a cube of edge 1 cm. This is
shown in the figures below, where the cuts are numbered and the rearranged pieces
are represented by dotted lines.
In Figure 1, the longest side of the triangle has length 6 cm. In Figure 2, one
of the shorter sides of the triangle has length 6 cm. In this case, it will be observed,
the third cut goes through two different pieces of the triangle. This may be consi-
dered a violation of the conditions of the problem, but I don't know how to get around
it. In Figure 3, the triangle is right-angled and one of the legs has length 6 cm.
Two cuts suffice in this case.

Figure 1 Figure 2 Figure 3

Also solved by F.G.B. Haskell, Algonquin Collegej and Leo Sauve, College
Algonquin.

25 i Proposed by Viktors Linis, University of Ottawa.

k I
Find the smallest positive value of 36 - 5 where k and I are positive
integers.
I. Solution by F.G.B. Maskell, Algonquin College.
We have
6 - 1 - 5 = ( 6 - 1)(6 + 1) - 5 ^ 0,
ll I
since 6 + 1 has at least one factor which is not 5, and 5 does not; hence
,,k X 2k X , „
36 - 5 = 6 - 5 f= } .
But the final digit of 36 - 5 must be 1, since the final digits of 36 and 5 are
6 and 5 respectively. Hence the minimum value is 11, which occurs when k = 1 and
1= 2.
II. Solution d'Andre Ladouceur, Ecole Secondaire De La Salle.
Les puissances de 36 se terminent par 16, 36, 56, 76, ou 96, i.e. elles
onfTaforme (2n + 1)1 o + 6. En effet, par induction, on voit que
36[(2n + 1)1 0+ 6] = {l2n + 57) 1 0 + 6 ,
 - i)3 -

et im + 57 est impair. On v o i t de meme que les puissances de 5 superieures a 5

f i n i s s e n t par 25, car

5(1 COn + 25) = (5M + 1)1 0 0 + 2 5 .

k I
La quantite 3 6 - 5 doit done finir par 91, 11, 31, 51, oD 71; elle ne peut
done etre inferieure a 11. Or elle prend la valeur 11 quandfe.= 1 et I = 2.
III. Solution by the proposer.
b p b p
Since 3 6 - 5 s 1 (mod 1 0) and 36 - 5 = 3 (mod 4 ) , we must have
fe t fe H
36 - 5 > 11; but forfe= 1 and I = 2, 36 - 5 = 1 1 , which gives then the required
minimum.
IV. Solution by L4o Sauv6, Algonquin College.
Since every power of 36 ends in 6 and every power of 5 ends in 5, the dif-
k t
ference 3 6 - 5 must end in 1. Forfe- 1 and I = 2, the difference is 11. This
will be the required value, unless we can find values of fe and I for which the dif-
ference is 1. Suppose such values of fe and I exist; then
36fe - 1 = 5l.
P
The l e f t side i s d i v i s i b l e by 36 - 1 = 5-7; hence 5 i s d i v i s i b l e by 7, which i s im-
p o s s i b l e . The answer i s 1 1 .
Comment by the -proposer.
Are there other values o f fe and I , besides fe = 1 and 1 = 2, which solve the
h P b P £
equation 36 - 5 = 11? The answer is no. For if 36 - 5 = 11, then (-1) = i(mod 6)
and t must be even, say 2m, and
fe
36 - 5 2 m = 62k - 52m = (6 fe - 5 m )(6 f e + S*) = 1 1 .

Since 11 i s a prime, 6 - 5m = 1 and 6 + 5m = 1 1 , so t h a t 6 = 6 , sm = 5, and fe = . 1 ,

m = 1 {I = 2) i s the only s o l u t i o n .

2 6 1 Proposed by Viktors Linis, University of Ottawa.

Given n i n t e g e r s . Show t h a t one can s e l e c t a subset o f these numbers and
i n s e r t plus or minus signs so t h a t the number obtained i s d i v i s i b l e by n.
I .Solution d'Andre" Ladouoeur, Eoole Secondaire De La Salle.
I I y a n classes d'equivalence modulo n . Si un des n nombres est equivalent
a 0 (mod n ) , i l est d i v i s i b l e par n. Dans l e cas c o n t r a i r e , i l d o i t y avoir deux des
nombres, disons a et 3, qui sont dans l a meme classe d'equivalence, et l e u r difference
a- B est a l o r s d i v i s i b l e par n.
II.Solution by Leo SauvS, Algonquin College.
Let the given integers b e a . , 1 = 1 , 2 , . . . , n , With appropriately inserted +
and — s i g n s , we form the sum ^

S„=b.+b,+ .... +6 (1)

where b . = \a.\. For k = 1, 2,..., n, let r, be the remainder obtained when S, is

divided by n. If r, = 0 for some k, then the corresponding S, is divisible by n.
Otherwise, since there are only n possible remainders, there must exist positive
 - w-
integers i and j , i < j , such that r. = r., and then
^ 3
S. - S. = fc. , + b. „ +...+ b .
is divisible by n. We have thus shown the interesting result that in any case there
exists a subset of consecutive terms of (1) whose sum is divisible by n.
Also solved by the proposer.

27. Propose peer L6o SauvS, College Algonquin.

Soient A, B, et C les angles d'un t r i a n g l e . I I est f a c i l e de v e r i f i e r que
si A = B = 4 ? , alors

cos A cos B + s i n A sin B s i n C = 1

La proposition reciproque e s t - e l l e vraie?

Solution by Viktovs Linis University of Ottawa.
For angles A, B, C i n a t r i a n g l e , we have, using the given c o n d i t i o n ,

1 = cos A cos B + sin A sin B sin C < cos A cos B + sin A s i n B = cos(A - B) < 1 .

Therefore the e q u a l i t y signs must hold throughout, and t h i s implies sin C = 1 and
cos(A - B) = 1. Hence C = 9 o° and A = B = 45°.
Also solved by G.D. Kaye Department of National Defence; Andre Ladouceur,
Ecole Secondaive De La Salle; F.G.B. Maskell, Algonquin College; and the proposer.

2 8 . Proposed by Le"o Sauve, Algonquin College.

If 1% of the population escapes getting a cold during any given year, how
many days must the average inhabitant expect to wait from one cold to the next?
I. Solution by the proposer.
We assume that the required time is n years from the start of one cold
to the start of the next, and that all times of year are equally likely for a cold
to begin.
Suppose a year is divided into k equal periods, where k is large, so that the
required time contains nk periods. The probability of escape in any one period is
approximately 1 - -=- , and the probability p of escaping for a whole year or k
periods is approximately (i - -^-) . Hence
_ j_
fe n
P=,l1m
fe -s-(1
00 - 4 )
nk' = e" .

If we equate this to o. 07 , we obtain

n = {In — — ) years = 137 days.

 - 45 -

II. Solution by G.D. Kaye, Department of Motional defence.

If it is assumed that the incidence of colds is random and independent,
the distribution of colds is Poisson, which is a one-parameter distribution and hence
convenient for this problem. The assumption is adequate only if the duration of each
cold is relatively short so that each individual may be exposed to a new cold several
times aa year.
ye Under the Poisson distribution, if the average number of colds per head
per year is x, the proportion of the population with k colds per year is given by
the term in a; of

f: +
C •X
(1 + x +
k\

In this case we have e 1 = 0.07, • giving x = 2.659. Thus the average man can expect
365
to wait = 137 days from the onset of one cold to the onset of the next.
2.659
There is insufficient information to determine the period from the end of one
cold to the start of the next.
Also solved by E.G. Dworschak, Algonquin College.

29 • Proposed by Viktors Linis, University of Ottawa,

Cut a square into a minimal number of triangles with all angles acute.
Solution by the proposer.
The answer is 8 for the square, and the same holds for any rectangle. (For
an obtuse triangle the answer is 7.) The following construction explains the proce-
dure for any rectangle ABCD. Draw two semicircles on AB and DC (longest sides) and
two semicircles on DM and MA, where M is the midpoint of AD (shortest side). The
shaded region gives the location of the "permissible" vertices P and q such that the
eight triangles as shown in the drawing are all acute, (ffl is the midpoint of CB.)
The shaded region is never an empty set: the point V where the two larger semicircles
intersect each other lies within the rectangle and outside the two smaller semicircles,
since it lies on MN.
 - 46 -

Editor's comments.
(i) The minimality of number 8 for the rectangle is a nearly immediate con-
sequence of the fact, mentioned by the proposer, that for a non-acute triangle the
minimal number is 7. For a proof of this fact, see American Mathematical Monthly,
Vol. 67 (i960), page 923.
(ii) A "permissible" area for locating vertices P and Q can also be found
by drawing semicircles on the shortest sides, and then drawing semicircles on the
two halves of both longest sides, as shown in the figures below where the "permissible
area is shaded.

Also solved by F.G.B. Haskell, Algonquin College, and Leo Sauve, College
Algonquin. One incorrect solution was received.

30. Proposed by Leo Sauve, Algonquin College.

Let a, b, and e denote three distinct integers and let P denote a polynomial
having all integral coefficients. Show that it is impossible that P(a) = b, P(b)=c,
and P(c) _= a. [Third VSA Mathematical Olympiad - May 7, 1974)

Solution by Viktors Linis, University of Ottawa.

A nice, short solution is given in The Mathematics Teacher, Vol. 68, No. 1
(Jan. 1975). It can hardly be improved.

Editor's comment.
The solution referred to above appeared in a report on the Third USA Mathe-
matical Olympiad written by Samuel L. Greitzer, Chairman of the Olympiad Committee.
There is insufficient space to reproduce this solution here in its entirety.
But The Mathematics Teacher is widely available, and interested readers should be
able to look it up there.
Also solved by G.D. Kaye, Department of National Defence; F.G.B. Maskell,
Algonquin College; and the proposer.

, Six is a number perfect in itself, and not because God created all things
in six days; rather the inverse is true, that God created all things in six days
because this number is perfect, and would remain perfect, even if the work of the